Small Subwoofer Drivers in Tower Speaker

[D!sco]

Hey all, I've been reading more and more about speaker design since I built my first set and I have an idea I'd hope to be able to execute in a few years.
The 4" Tang Band midbass drivers I have on my desk are impressing the hell out of me with their low end. Tang Band clearly knows how to produce a lot of subsonic with their smaller drivers. I've also been a fan of the 6.5" driver width from an aesthetic standpoint. It's a nice size and can hide box depth reasonably well. I'm willing to bet an all 6.5" DIY tower would look quite nice done right. My current inspiration is the KEF R7.

Finally, the drivers and crossover points I had in mind were pretty simple, either a 6.5" coaxial or an equally sized midbass driver and a good tweeter. I like the idea of going coaxial and the W6-2313 has some good examples/reviews, same with the MR18REX/XF. From what I understand, coaxials do better when they are at lower excursion, so I would want to give it a generous (sealed?) box and an early high pass filter. Could a well-regarded, small diameter sub, such as the W6-1139SI actually come up to meet a driver at, say, 400hz and sound good? I'm not trying to replace a good home theater sub, but I wonder if I could hit 30-450hz flat with one or two of these and some separate power input.

What are some of the pitfalls of a design like this? I could just as easily put normal 6.5" woofers to do the same job and get a sub or two later. Graphs look great and all, but I don't know if it would be "fast" enough to play nicely. The goal of a coaxial driver would be to have nice dispersion and off-axis, too. Would a smaller or wider baffle be preferable for this?

 

[Wolf]

I used the 1139 to about 650, and it rolls off hard there. The smaller 1138 is good to about 1.1kHz.

So, yes- they should suffice. You may have to pad the coax or separately amplify the 1139 or other to get the same output as the coax and get a good blending.

 

[Rick Sykora]

Have to be careful with Tang Band. They have some nice drivers, but the scaling on their specs is too wide AND they do not publish off-axis response. Does not exactly instill high confidence in their drivers.

If you found some more info on the coax, please post a link. Thanks!

 

[Wolf]

A lot of people have said this, and while I agree that the FR is too smooth and does not show the larger aberrations, I have had great results using their drivers' response curves in design preparations. If you want off-axis responses, you can e-mail them. I have seen some they have taken and supplied to builders in the past.

I will say this though- a lot of their woofer drivers tend toward really low sensitivity, and this severely limits usage in a lot of cases. There are also drivers of atypical design mantras that really aren't something I would use. Then there are the drivers that are really close to state of the art, and they test really well.

TB is a company that you have to pick and choose your risk factors, but more often than not come out sounding rather good.

W6-2313 link at diyaudio, where perceval does his homework in his design stages:
https://www.diyaudio.com/forums/multi-way/324159-tb-line-coax-fr-drivers.html

 

[D!sco]

Thanks for the replies! I was just about to post some work I'd done on enclosures and potential pitfalls I'd found.

I have an enclosure designed with appealing overall dimensions for living room use. The question is whether I seal half of the drivers and vent the other half for better midrange response or run all two/four in the same enclosure. Purple circle indicates optional chamber separation.

The plan is to add a 4" flared port to the front, rear, or downfiring.
EDIT: The top chamber would be 20.1L and the bottom would be 84.5 minus the 4x5.5" port. The full box comes to 104L with a 4"x6.5" port

Some graphs


Seems like the combination has a smoother response into the midbass range, but all of them have phase problems(?)
It's a good problem to pick at while things are returning to normal.

 

[Wolf]

Vent them. Those woofers need to breathe, and if you run them sealed they do not get the extension you will so desire. If they all are operating in the same structure, they can be one volume and one tuning, or they can be split and have the same tuning. Multiple exit tunings messes things up.

If you want to try out multi-chamber tunings, try this interface:
https://www.micka.de/en/advanced.php

 

[Rick Sykora]

A lot of people have said this, and while I agree that the FR is too smooth and does not show the larger aberrations, I have had great results using their drivers' response curves in design preparations. If you want off-axis responses, you can e-mail them. I have seen some they have taken and supplied to builders in the past.

I will say this though- a lot of their woofer drivers tend toward really low sensitivity, and this severely limits usage in a lot of cases. There are also drivers of atypical design mantras that really aren't something I would use. Then there are the drivers that are really close to state of the art, and they test really well.

TB is a company that you have to pick and choose your risk factors, but more often than not come out sounding rather good.

W6-2313 link at diyaudio, where perceval does his homework in his design stages:
https://www.diyaudio.com/forums/multi-way/324159-tb-line-coax-fr-drivers.html

Not a short post (~3 years worth), so tend to read in reverse chronological order. One of perceval's last posts says his w6-2313s died and "might want to wait for the MkII".

 

[D!sco]

I've sent an e-mail out to TB asking for off-axis responses of their coaxials. It would be nice if they listed separate tweeter specs for power handling as well. Maybe I can get more info if they respond.

 

[D!sco]

W6-2313 link at diyaudio, where perceval does his homework in his design stages:
https://www.diyaudio.com/forums/multi-way/324159-tb-line-coax-fr-drivers.html

One of the most informative reads on that driver I've found. Worth the registration to get all the pictures, too. The ending is a bit of a shame with no follow up on the tweeter fade. At least there are still good alternatives.

 

[Biblob]

For a good coaxial driver, take a look at the offerings from Sica.
They are shortly discussed here:
https://www.audiosciencereview.com/...orm-speaker-project.20407/page-32#post-878844

The fact that @sigbergaudio uses the 5,5" should speak for something . Maybe @sigbergaudio would be willing to share the raw measurements from this driver?

 

[Biblob]

How about using a bigger subwoofer on the side of the box? Take a look at the Infinity Il50 and the likes. That will definitely give you the possibility to have ~30hz flat.

 

[D!sco]

How about using a bigger subwoofer on the side of the box? Take a look at the Infinity Il50 and the likes. That will definitely give you the possibility to have ~30hz flat.

That was what I was thinking originally, build a standard TMM with side subs. Part of the goal is visual. It's a lot to have five drivers pointed at you at once. The other is to reduce my crossover by (hopefully) utilizing some high quality, smaller subs as midbass woofers as well. If I scale down to the 5.5" Sica, I could easily integrate a smaller and more agile woofer/sub like the Tang Band W5-1138SMF, so a different Coaxial entirely isn't out of the question.

Last, SB Acoustics actually has a set of matching coaxial drivers and woofers with an Xmax of 9mm and a resonant frequency of 38Hz. It's a much more affordable speaker to practice on, and I have the premise of the Buchardt A700 for stacking these exact woofers. When I plug the values into VituixCAD, they look easy to EQ and inexpensive at $412 for 10 drivers. They even list the crossover points on their site for making a four way out of those woofers and a Satori(?) tweeter. 120Hz, 250Hz, and 2800Hz. It looks decent when plugged into VituixCAD, too. (Tweeter is inverted)

 

[D!sco]

For a good coaxial driver, take a look at the offerings from Sica.
They are shortly discussed here:
https://www.audiosciencereview.com/...orm-speaker-project.20407/page-32#post-878844

The fact that @sigbergaudio uses the 5,5" should speak for something . Maybe @sigbergaudio would be willing to share the raw measurements from this driver?

If the specs are to be believed it's SEAS quality at Dayton pricing. I like the idea, a lot. I wish I had more to go on. They look really promising. Thanks for sharing.

 

[sigbergaudio]

If the specs are to be believed it's SEAS quality at Dayton pricing. I like the idea, a lot. I wish I had more to go on. They look really promising. Thanks for sharing.

They measure as specified and can also take a lot of power and abuse as long as you roll them off early enough.

The speaker we've built based on this driver is +/-2dB from 100-20,000hz.

Here's a recent measurement from 1khz-20khz 15 degrees off-axis (1/6 smoothing):

And more importantly, they also sound fantastic of course!

 

[D!sco]

They measure as specified and can also take a lot of power and abuse as long as you roll them off early enough.

That's really good to know. They cost 2/3rds as much as the SEAS and Tang Band offerings as well.

If you found some more info on the coax

Follow up on the Tang Band drivers, they got back to me last night with off-axis response.

 

[Rick Sykora]

@D!sco while nice that TB supplied off-axis info, note the y-axis scale is twice the range of the one posted by sigbergaudio.

So, is not near as flat on any axis compared to others that post on the more typical scale.

 

[sigbergaudio]

@D!sco while nice that TB supplied off-axis info, note the y-axis scale is twice the range of the one posted by sigbergaudio.

So, is not near as flat on any axis compared to others that post on the more typical scale.

Please note that my graph is the corrected/finished response of our speakers, not raw driver response.

 

[D!sco]

@Rick Sykora Yeah, I noticed that tracing just the on-axis into vituixCAD last afternoon. Those little lines are gulfs in a 30dB scale.

@sigbergaudio, do you have distortion/off-axis measurements for the 5.5” or 6.5” woofer? I’d like to know where to cross optimally in a 3-way configuration. By the way, I saw your bookshelf and it looks fabulous.

 

[sigbergaudio]

@Rick Sykora Yeah, I noticed that tracing just the on-axis into vituixCAD last afternoon. Those little lines are gulfs in a 30dB scale.

@sigbergaudio, do you have distortion/off-axis measurements for the 5.5” or 6.5” woofer? I’d like to know where to cross optimally in a 3-way configuration. By the way, I saw your bookshelf and it looks fabulous.

I don't have raw driver measurements readily available. For a 3-way around 400hz is probably a good start.

 

[Rick Sykora]

@Rick Sykora Yeah, I noticed that tracing just the on-axis into vituixCAD last afternoon. Those little lines are gulfs in a 30dB scale.

@sigbergaudio, do you have distortion/off-axis measurements for the 5.5” or 6.5” woofer? I’d like to know where to cross optimally in a 3-way configuration. By the way, I saw your bookshelf and it looks fabulous.

TB did not happen to mention how much smoothing may have been applied to the measurements?

Should be designing with no more than 1/24 btw.